ModelArts

The ModelArts connector interconnects with Huawei Cloud ModelArts.

ModelArts is a one-stop AI development platform. It provides data preprocessing, semi-automated data labeling, large-scale distributed training, automated model generation, and model deployment on devices, edge devices, and the cloud. It helps you develop and deploy models quickly and manage the AI development lifecycle for machine learning and deep learning.

Creating a ModelArts Connection

Log in to the new ROMA Connect console.
In the navigation pane on the left, choose Connector. On the page displayed, click New Connection.
Select the ModelArts connector.

In the dialog box displayed, configure the connector and click OK.

Parameter	Description
Name	Enter the connector instance name.
App Key	Access key ID (AK) of the current account. Obtain the AK by referring to Access Keys. If an AK/SK pair has been generated, find the downloaded AK/SK file (such as credentials.csv).
App Secret	Secret access key (SK) of the current account. Obtain the SK by referring to Access Keys. If an AK/SK pair has been generated, find the downloaded AK/SK file (such as credentials.csv).
Description	Enter the description of the connector to identify it.

Action

Deleting nodes in batches
Listing all nodes
Obtaining the real-time resource usage
Resource pool statistics
Resource pool monitoring
Deleting a resource pool
Querying a resource pool
Creating a resource pool
Listing all resource pools

Configuration Parameters

**Table 1** Deleting nodes in batches
Parameter	Description
project_id	Project ID.
pool_name	Resource pool name.
region_id	Region ID.
deleteNodeNames	Names of the nodes to be deleted.

**Table 2** Listing all nodes
Parameter	Description
project_id	Project ID.
pool_name	Resource pool name.
region_id	Region ID.
continue	Position of the previous page during pagination query.
limit	Number of records returned for a single pagination query.

**Table 3** Obtaining the real-time resource usage
Parameter	Description
project_id	Project ID.
region_id	Region ID.

**Table 4** Resource pool statistics
Parameter	Description
project_id	Project ID.
region_id	Region ID.
workspaceId	Workspace. The default value is 0.

**Table 5** Resource pool monitoring
Parameter	Description
project_id	Project ID.
pool_name	Resource pool name.
region_id	Region ID.
time_range	Time range for query. The default value is -1.-1.60. The format is startTimeInMillis.endTimeInMillis.durationInMinutes. Parameter description: startTimeInMillis: start time (UTC) of the query, in milliseconds. If this parameter is set to -1, the server calculates the start time as follows: endTimeInMillis – durationInMinutes × 60 × 1,000. endTimeInMillis: end time (UTC) of the query, in milliseconds. If this parameter is set to -1, the server calculates the end time as follows: startTimeInMillis + durationInMinutes × 60 × 1,000. If the calculated end time is later than the current system time, the current system time is used. durationInMinutes: query time span, in minutes. The value must be greater than 0 and no less than the result of (endTimeInMillis – startTimeInMillis)/(60 × 1000) – 1. If both the start time and end time are set to -1, the system sets the end time to the current UTC time (in milliseconds) and calculates the start time as follows: endTimeInMillis – durationInMinutes × 60 × 1,000. For example, -1.-1.60 indicates the latest 60 minutes. Constraint: In a single request, the following condition must be met: durationInMinutes × 60/period ≤ 1,440
statistics	Statistical mode. The options are: maximum (default) minimum sum average sampleCount
period	Monitoring data granularity, in seconds. The options are: 60 (default): 1 minute 300: 5 minutes 900: 15 minutes 3600: 1 hour

**Table 6** Deleting a resource pool
Parameter	Description
X-ModelArts-User-ID	Actual external tenant ID. If there is a tenant ID, it will be used for workspace authentication.
project_id	Project ID.
pool_name	Resource pool name.
region_id	Region ID.

**Table 7** Querying a resource pool
Parameter	Description
X-ModelArts-User-ID	Actual external tenant ID. If there is a tenant ID, it will be used for workspace authentication.
project_id	Project ID.
pool_name	Resource pool name.
region_id	Region ID.

**Table 8** Creating a resource pool
Parameter	Description
X-ModelArts-User-ID	Actual external tenant ID. If there is a tenant ID, it will be used for workspace authentication.
project_id	Project ID.
region_id	Region ID.
apiVersion	API version. Option: v2
kind	Resource type. Option: Pool
metadata	Resource pool metadata.
labels	Resource pool labels.
os.modelarts/workspace.id	Workspace ID. The default value is 0.
os.modelarts/name	Specified resource pool name.
annotations	Resource pool annotations.
os.modelarts/description	Resource pool description.
os.modelarts/billing.mode	Billing mode. Options: 0: pay-per-use 1: yearly/monthly
os.modelarts/period.num	Subscription period on a yearly/monthly basis, for example, 2. This parameter is mandatory when the billing mode is yearly/monthly.
os.modelarts/period.type	Subscription type on a yearly/monthly basis. Options: 2: monthly 3: yearly This parameter is mandatory when the billing mode is yearly/monthly.
os.modelarts/auto.renew	Whether to enable auto-renewal. Option: 0 (default): Auto renewal is disabled.
os.modelarts/promotion.info	Discount selected in CBC.
os.modelarts/service.console.url	URL of the page displayed after the subscription order is paid.
os.modelarts/order.id	Order ID, which is mandatory for creating a yearly/monthly resource pool or changing the billing mode
spec	Resource pool description.
type	Resource pool type. Options: Dedicate: physical resource pool, which uses a separate network and supports network connection, custom drivers, and custom job types.
scope	Job type supported by a resource pool. Select at least one type. All options are supported by physical resource pools. Options: Train Infer Notebook
resources	Resource flavor list in a resource pool, including resource flavors and the number of resources for each flavor.
flavor	Resource flavor, for example, modelarts.vm.gpu.t4u8
count	Number of resources of the specified flavor
network	Network settings for a resource pool. This parameter is mandatory for physical resource pools.
name	Network name, that is, the value of the metadata.name field in the network details. When you create a network with a specified name, the system will automatically create subnets for you. By default, the first subnet will be used.
jobFlavors	Names of the job flavors supported by a resource pool.
driver	Resource pool driver information.
gpuVersion	GPU driver version. This parameter is available when GPUs are used in a physical resource pool. For example, the GPU driver version is 440.33.
npuVersion	NPU driver version. This parameter is available when Ascend chips are used in a physical resource pool. For example, the Ascend driver version is C78.
updateStrategy	Driver upgrade policy. Options: force: forcible upgrade. The node drivers are upgraded immediately, which may affect jobs running on the node. idle: secure upgrade. The drivers are upgraded when no job is running on the node.

**Table 9** Listing all resource pools
Parameter	Description
project_id	Project ID.
region_id	Region ID.
workspaceId	Workspace ID. The default value is 0.
labelSelector	Label filtering.
status	Resource pool status. Options: created: The resource pool is successfully created. failed: The resource pool failed to be created. The failure record is retained for three days. creating: The resource pool is being created. If this parameter is left blank, valid resource pools are returned, including those being created and those created.